Feature Request - Enterprise Manager - Fix This

[ChuckS42]

Regarding bullet point 11 on this document: https://helpcenter.veeam.com/docs/backu ... tml?ver=95

We use storage array replication to copy our production backup data to our secondary datacenter, then import this data in to a separate B&R server (by creating read-only repositories). We have one Enterprise Manager server covering both locations (primarily for managing licenses). We suffer the bad stuff from the above document. I would very much like to see this resolved, or at least give me the option of ignoring the replicated repositories (since the remote B&R server has backups of its own).

(I had a case open for the database errors; # 02719203.)

[PTide]

Hi Chuck,

May I ask you what is the purpose of importing the copies into a separate VBR server? Do you perform some post-replication actions with the imported backups? Also, would you tell what storage array do you use please?

Thanks

[ChuckS42]

Our secondary datacenter is our DR site. The purpose of the replication is so I can quickly restore backups in case the primary datacenter goes down in a disaster. No post-replication actions. My current backup storage is a pair of Quest (formerly Dell) DR6000 arrays. We are looking at replacing them, and have talked with Exagrid; they actually recommend this method of replicating Veeam backups and were surprised when I told them of this Enterprise Manager limitation.

I am aware I could use the Backup Copy functionality of Veeam to accomplish this goal, but it is my preference to allow the storage array to handle site-to-site replication, as it is more efficient and simpler to configure/manage.

[PTide]

The purpose of the replication is so I can quickly restore backups in case the primary datacenter goes down in a disaster

Yes, I understand, but I meant to ask why don't you just replicate them to the DR without actually importing them? In case a disaster happened you would detach your primary VBR (no use of it since it's on the dead site) from EM and create a new repository of the folder where the replicated backups are stored. After that you would import them to the second VBR and use for restore. I agree that this is just a workaround, but it should work until we fix the limitation, unless I'm missing something.

Thanks

[ChuckS42]

No, that makes sense. It just adds a bunch of extra steps to a DR plan. Better to have it all in place in advance.

[PTide]

I agree, and now I'm going to say a horrible thing: in order to slightly reduce the number of steps you can just ignore the 'detach' part and go straight to importing backups. That will screw up EM database just the same way it does with your current approach, but I guess that won't be your main concern during DR situation and can be sorted out later when everything is up and running.

Anyway, your request has been noted.

Thanks

[ChuckS42]

Thank you.

[ChuckS42]

Yay necropost!

Question - When I installed Update 4 last month, it appeared that this was no longer an issue (the Machines view in EM showed both sets of backups, original and replicated). However, for reasons I needed to remove and reinstall EM with a new database, and after re-adding the Backup servers and performing the database replication, the old "duplicate key row" errors are back. Was I imagining things before?

[PTide]

Unfortunately, the following limitation still holds true:

You cannot add backup servers that process same machines.
You cannot add backup servers that process same backups.

Since you mentioned array replication, I suspect that it is exactly what clauses 2 says - you cannot have servers that process the same backups/VMs added into EM. I'll try to find out why it sort of 'worked' after updating to U4. Whatever the reason for that, such scenario is not supported yet.

Thanks!

[ChuckS42]

Ah, too bad. I was tripped up by the fact that your Update 4 doc for EM omitted the "Considerations and Limitations" section from the previous versions of the doc. (Current URL: https://helpcenter.veeam.com/docs/backu ... l?ver=95u4).

When I say that it worked, I mean that after I ran the Update 4 upgrade, I went into the EM website and clicked on the Machines tab, I saw two entries for each machine - one for the original backup that I could interact with, and a second entry linked to the remote Backup Server that I couldn't.

As I mentioned originally, this is not a game-ender, because we don't use EM for restores. But it would be nice. I can't believe I'm the only person who uses replicated repositories; Exagrid, for example, is designed for this purpose (and they were surprised this problem exists in the first place).

[jim3cantos]

Regarding bullet point 11 on this document: https://helpcenter.veeam.com/docs/backu ... tml?ver=95
...
I would very much like to see this resolved, or at least give me the option of ignoring the replicated repositories (since the remote B&R server has backups of its own).

+1 to this FR. Enterprise Manager isn't "compatible" with the dummy job workaround.